NRDF: Neural Riemannian Distance Fields for Learning Articulated Pose Priors

Manifold of Plausible Articulated Poses Quaternion Geodesic Distance P r o d u c t M an if ol d of Riemannian Qu at er ni o n s Figure 1. Left: We present Neural Riemannian Distance Fields (NRDFs), a principled method to learn data-driven priors as subspace of high-dimensional Riemannian manifolds. Right: NRDFs can effectively model the pose of different articulated shapes. We present diverse samples generated using NRDFs trained on human, hand, and animal poses respectively.